At an internal meeting held on Wednesday, August 19, OpenAI Chief Financial Officer Sarah Friar presented a clear yet deliberately ambiguous picture to the company's workforce about the organization's near-term future. The central message — repeated across slides and spoken remarks — was that OpenAI "will be a public company in 2027," or sooner if growth holds. The disclosure, relayed by CNBC to unauthorized sources, consolidated a direction that had been taking shape behind the scenes for months.
In June, The New York Times had reported that OpenAI was considering pushing its initial public offering (IPO) into 2027 rather than listing the company in late 2026. At the time, three people involved in the deliberations stated that the board was leaning toward waiting, although CEO Sam Altman preferred a Q4 2026 listing. The internal conflict between Friar's cautious posture and Altman's impatience was genuine: the CFO argued the company needed robust financial controls before opening its capital to the global market, while Altman saw a rapid listing as an opportunity to capitalize on product momentum.
So what has changed?
The CFO's presentation revealed striking numbers. OpenAI's revenue run rate grew 35% quarter-to-date — the "run rate" is a financial metric that projects annual revenue based on current quarterly performance. Enterprise revenue, meaning income from corporate contracts and B2B licenses, surged 50% over the same period. These figures suggest a structural transformation: at the end of 2024, approximately 75% of OpenAI's revenue came from consumer ChatGPT subscriptions. Today, the enterprise base represents a majority share, and this transition is precisely the type of metric that Wall Street investors seek — recurring, predictable revenue, with less dependence on a single consumer product.
The second-quarter 2026 revenue figures put the growth into perspective. According to the Wall Street Journal, OpenAI generated $6.7 billion in Q2 revenue, up from $5.7 billion in the first quarter. The annualized run rate now exceeds $40 billion, meaning the company has doubled its annual revenue run rate compared to the end of 2025, as reported by Bloomberg. For comparison, a company generating $40 billion in annual revenue is at the level of technology giants. The problem is that OpenAI is still not profitable — and the gap between revenue and profit is the most alarming dimension of the balance sheet.
The 2026 loss projection hovers around $14 billion, nearly triple the 2025 loss of approximately $20.9 billion in operating losses. The reason is simple and unforgiving: compute costs. Just from Azure spending and data center infrastructure, OpenAI moves about $13 billion this year. The payroll of approximately 4,500 employees costs about $4 billion annually. The company spends approximately $1.69 for every dollar it earns — a burn rate that requires constant fresh capital.
This financial context explains the urgency of the IPO. Companies burning $1.5 billion per month need capital markets to sustain operations. OpenAI is no exception. But the timing is not simple: listing a company with such significant losses in such a volatile sector means that analysts will scrutinize every boardroom decision through the lens of the market. A strategic misstep, a delay in launching a model, a loss of market share to a competitor — all of this could tank the stock price in the first days of trading.
And here enters the third critical element: the rivalry with Anthropic. In Q2 2026, Anthropic surpassed OpenAI in quarterly revenue for the first time, with $11.6 billion versus OpenAI's $6.7 billion. The gap is significant. While OpenAI focuses on ChatGPT and API infrastructure for third parties, Anthropic has advanced with enterprise integration strategies — Claude is being incorporated into tools at companies like Salesforce, Shopify, and other large-scale platforms. The quarterly revenue difference between the two companies now reaches nearly $5 billion. If Anthropic maintains this pace, the distance between the two competitors will only widen.
Even Friar urged calm on this aspect. During her presentation, she directed employees not to worry about Anthropic's timeline — the competitor also has been pressing for a 2027 listing. But this calm may be strategic: Anthropic already operates with even deeper losses (estimates project $19 billion in losses for 2026) and depends on external investors to sustain operations.
What emerges from this panorama is an inflection point in the artificial intelligence sector. The two largest companies in the industry — OpenAI and Anthropic — are both seeking capital markets in the same year, amid regulatory uncertainty about frontier models and growing investor pressure for profitability. OpenAI's IPO will be the sector's largest — estimates point to a valuation of $852 billion to $1 trillion — and the coming weeks will determine whether the stock market is prepared to absorb a company of this scale during a period of macroeconomic volatility.
The question that remains is whether OpenAI truly needs the IPO to survive or whether it is, first and foremost, a release valve for internal investors who need to realize gains — the company itself had reportedly bought back $7 billion worth of employee shares at an $852 billion valuation. In either case, the clock has already started ticking.
Sources: CNBC, The Next Web, Crypto Briefing
✓ Independent sources cross-checked and verified before publishing
